Type 2 diabetes mellitus (DM2), is causally associated with coronary artery disease (CAD), and is a major health and economic burden. Despite multiple preventive and therapeutic measures implemented in the last few decades to decrease the occurrence and progression of cardiovascular (CV) disease in DM2, at the population level these measures remain insufficient. Current strategies addressing the epidemic of DM2 and CV disease may be greatly enhanced by the identification of novel pathophysiological mechanisms. Adipose tissue is a bioactive organ that releases numerous paracrine and endocrine mediators (adipokines). Changes in adipokine levels may contribute to metabolic and cardiovascular consequences of DM2, and may also help explain any association between cardiovascular prognosis and various therapeutic strategies in DM2. Effects of treatment may in part be related to changes in plasma adipokine profiles. However, the role of adipokines in the pathophysiology of cardiovascular disease in DM2 has not been well established. We propose a series of novel studies directed at investigating the association between adipokines and DM2, with respect to severity of associated CAD, cardiovascular prognosis, effects of therapy, and risk stratification. The present proposal is intended to be an ancillary study to the ongoing prospective Bypass Angioplasty Revascularization Investigation 2D (BARI 2D) trial. As part of the BARI 2D protocol, complete data regarding demographics, clinical characteristics, traditional risk factors, angiographic findings, and medical treatment will be available in 2368 patients at baseline and at regular intervals, during a 5-year follow-up period. We will test the following primary hypotheses: 1) That treatments with insulin provision versus insulin sensitizing drugs exert differential longitudinal effects on the adipokine levels (or their principal components) in DM2. 2) That baseline levels and changes in adipokine levels (or their principal components) over time are predictors of CV outcome in DM2. 3) That the effect of the randomized treatment on CV outcome in DM2 is mediated (or contributed to) by adipokine levels (or their principal components). Identification of specific pathophysiological factors involved in the development, progression, and prognosis of CAD in individuals with DM2, and their responses to different treatment options, will enable more optimal therapeutic strategies to be designed.